OpenAI and Leading AI Labs Commit to AI Governance

2023/07/21

This article was written by an AI 🤖. The original article can be found here. If you want to learn more about how this works, check out our repo.

OpenAI, along with other prominent AI labs, is taking significant steps to reinforce the safety, security, and trustworthiness of AI technology and services. These voluntary commitments, coordinated by the White House, aim to advance effective AI governance globally.

As part of their mission to build safe and beneficial AGI (Artificial General Intelligence), OpenAI will continue to pilot and refine concrete governance practices tailored to highly capable foundation models. They are committed to investing in research that can inform regulation, including techniques for assessing potentially dangerous capabilities in AI models.

The announcement of these commitments contributes to the ongoing collaboration between OpenAI, governments, civil society organizations, and others worldwide to advance AI governance. Policymakers are actively considering new laws for highly capable AI systems, and these commitments provide specific and practical practices for that discussion.

The voluntary commitments cover a range of areas, including safety, security, and transparency. Companies making these commitments agree to conduct internal and external red-teaming of models or systems to assess misuse, societal risks, and national security concerns. This rigorous evaluation process is crucial for building successful products, ensuring public confidence in AI, and mitigating significant threats.

It is important to note that these commitments apply to generative models that surpass the current industry frontier in terms of overall power. For example, they include models more powerful than GPT-4, Claude 2, PaLM 2, Titan, and DALL-E 2 in the case of image generation.

OpenAI's Vice President of Global Affairs, Anna Makanju, emphasizes the collaborative nature of these commitments, stating, "This announcement is part of our ongoing collaboration with governments, civil society organizations, and others around the world to advance AI governance."

By voluntarily committing to these practices, OpenAI and other leading AI labs are taking a proactive approach to ensure the responsible development and use of AI technology. These commitments will remain in effect until regulations covering similar issues are established.